A crossing predictor (often referred to as a grade crossing predictor in the U.S. or a level crossing predictor in the U.K.) is an electronic device which is connected to the rails of a railroad track and is configured to detect the presence of an approaching train and determine its speed and distance from a crossing (i.e., a location at which train tracks cross a road, sidewalk or other surface used by moving objects), and use this information to generate a constant warning time signal for control of a crossing warning device. A crossing warning device is a device which warns of the approach of a train at a crossing, such as crossing gate arms (e.g., the familiar black and white striped wooden arms often found at highway grade crossings to warn motorists of an approaching train), crossing lights (such as the two red flashing lights often found at highway grade crossings in conjunction with the crossing gate arms discussed above), and/or crossing bells or other audio alarm devices. Crossing predictors are often (but not always) configured to activate the crossing warning device at a fixed time (e.g. 30 seconds) prior to an approaching train arriving at a crossing.
Typical crossing predictors include a transmitter that transmits a signal over a circuit formed by the rails of the track and one or more shunts positioned at desired approach distances from the transmitter, a receiver that detects one or more resulting signal characteristics, and a logic circuit such as a microprocessor or hardwired logic that detects the presence of a train and determines its speed and distance from the crossing. The approach distance depends on the maximum allowable speed of a train, the desired warning time, and a safety factor. Preferred embodiments of crossing predictors transmit generate a constant current AC signal, and the crossing predictor detects a train and determines its distance and speed by measuring impedance changes due to the train's wheels and axle acting as a shunt across the rails and thereby effectively shortening the length (and hence the impedance) of the rails in the circuit. Those of skill in the art will recognize that other configurations of crossing predictors are possible.
It should be understood that trains are sometimes expected to move in both directions along a track. In such situations, a shunt may be placed at the desired approach distance on both sides of a crossing. Crossing predictors typically detect a train on either side of the crossing and activate a warning device when a train approaches from either direction, but do not have the ability to determine the direction of travel of a train along the track or distinguish a train on one side of the crossing from a train on the other side of the crossing (in other words, the crossing predictor can determine that a train is moving toward or away from it, but cannot determine from which side of the crossing the train is approaching). Such crossing predictors are sometimes referred to as bidirectional crossing predictors.
In certain locations, two or more crossings may be located within a desired approach distance of each other. In order to prevent the signals transmitted by one crossing predictor from interfering with another crossing predictor in such situations, the crossing predictors are often configured to transmit on different frequencies. This technique works well when the number of adjacent crossings is small. However, when the number of adjacent crossings gets larger, a problem can occur. A certain amount of separation between transmitted frequencies is necessary in order to ensure that a crossing predictor can reliably discriminate between its frequency and an adjacent frequency, and the maximum distance at which a train may be reliably detected is inversely proportional to the transmission frequency. Thus, only a certain number of unique frequencies at which the crossing predictors may transmit are available. Indeed, in some areas (particularly urban areas), not enough unique frequencies may be available to accommodate a number of crossings in close proximity with desired approach distances.
In order to address such situations, techniques for using a crossing predictor to detect and predict the arrival of a train at a downstream crossing and transmit a constant warning time signal to a device at the downstream crossing accordingly (i.e., generate and transmit a signal to activate the warning device at the downstream location when the speed and distance of a train are such that the train will reach the downstream crossing within a desired constant warning time). A term commonly used in the railroad industry for such prediction and signaling is “DAXing.” “DAX” is an acronym for “downstream adjacent crossing.” Further background information regarding DAXing can be found in U.S. Pat. No. 7,575,202, the contents of which are hereby incorporated herein by reference. It should be understood that the DAX signal may be transmitted by any means, including by radio or over a buried lines or above-ground wires.
Those of skill in the art will recognize that, for tracks on which trains may move in either direction, DAXing may be desired when a train moves in one direction but not in the other direction. For example, on a track running from east to west, it is desirable for a crossing predictor at a first crossing to DAX a second device at a nearby second crossing located to the east of the first crossing if a train is approaching the first crossing from the west. However, having the crossing predictor at the first crossing DAX the device at the second crossing may not be desirable in the event that the train were approaching the first crossing from the east.
In situations in which three (or more) crossings are closely located and a sufficient number of unique transmission frequencies are not available, it has been known to configure outer crossing predictors to DAX the inner crossing predictors (and, sometimes, to also DAX the downstream outer predictor). Because bidirectional crossing predictors cannot determine from which side of a crossing a train is approaching, and because it is desirable for an outer crossing predictor to DAX an inner crossing predictor only when the inner crossing predictor is downstream with respect to the direction in which a train is traveling, the outer predictors are made to act as unidirectional predictors by placing an insulated track joint at the location of the outer predictor. The insulated track joint only allows the transmitted signal to propagate in one direction along the track. The crossing predictor will employ two circuits, one on each side of the insulated joint, with each circuit therefore detecting trains on only one side of the crossing. The crossing predictor is equipped with logic that can determine whether the train in one circuit has previously been seen by the other circuit and therefore can DAX in only the desired direction. In other variations, insulated joints have been used in other ways to allow reuse of frequencies in dense areas.
The use of insulated track joints to accommodate crossing predictors as discussed above is costly, both in terms of the cost of initial installation and maintenance of the insulated track joints themselves, and in the need for additional changes to the installed signaling system, such as the need for coded track repeater units and filters.